© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 10121
    Prev
    Next
    INDEX
    Explanations

    websites or domain-related terms

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    365
    +0.10
    0.5%
    357
    +0.10
    0.4%
    527
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.10
    0.03
    365
    +0.10
    0.01
    527
    +0.10
    0.01
    Negative Logits
     suspic
    -0.86
    uate
    -0.76
    uated
    -0.69
     Reconstruction
    -0.64
    senal
    -0.60
     releg
    -0.59
    ikuman
    -0.58
     distracted
    -0.56
    istic
    -0.56
     culp
    -0.56
    POSITIVE LOGITS
    ritten
    1.30
    riter
    1.25
    estern
    1.14
    esome
    1.09
    een
    1.09
    atcher
    1.08
    ORD
    1.08
    ITNESS
    1.05
    esley
    1.02
    orld
    1.00
    Activations Density 0.066%

    No Known Activations